On August 5, 1949, fifteen smokejumpers parachuted into Mann Gulch above the Missouri River in Montana. Within about two hours of landing, thirteen men were dead: twelve of the jumpers and a Forest Service fire guard who had joined them on the ground. The crew’s only radio was destroyed in the drop because its cargo parachute failed to open. This piece covers the communications equipment a 1949 crew had and lacked, what the record supports about the missing radio and what it does not, and how this fire and the ones after it shaped the communications rules wildland crews carry today.

Thirteen dead, twelve of them jumpers: getting the count right

The death toll at Mann Gulch is thirteen, and the reputable accounts do not dispute it. Fifteen smokejumpers flew out of Missoula that afternoon under foreman R. Wagner Dodge, who went by Wag. On the ground they met James Harrison, the Forest Service fire guard stationed at the Meriwether campground on the river, who had been working the fire alone before the airplane arrived. That put sixteen men in the gulch, and because only Dodge and two jumpers, Robert Sallee and Walter Rumsey, survived, the thirteen who died were twelve smokejumpers and Harrison. According to Norman Maclean’s book and the Forest Service accounts, two of the injured men were found alive that evening and died of their burns the next day, and they are counted among the thirteen.

The figure is misreported in a few predictable ways, and each comes from attaching a real number in the story to the wrong group. “Thirteen smokejumpers died” is the most common error, and it is wrong because one of the thirteen was a fire guard who did not jump. “Twelve died” appears when a writer counts only the jumpers and drops Harrison. Fifteen or sixteen shows up when a writer confuses the number who jumped, or the number on the ground, with the number killed. Mann Gulch also gets mixed up with the South Canyon fire on Storm King Mountain in Colorado in 1994, where fourteen firefighters died, and that figure sometimes drifts into retellings of the 1949 fire.

For anyone who teaches this fire, getting the count right is part of the content of the lesson. Harrison joined the crew at the scene and was never on the jump manifest, so a count taken from the airplane’s roster would have missed him. Accountability systems on today’s firegrounds have the same weakness when they count from the dispatch record instead of from the people who are actually present. The memorial markers on the slope, one for each man where he fell, are the physical record, and Maclean’s book names all thirteen.

August 5, 1949: what the record supports about the afternoon

The fire started from lightning on the Helena National Forest, in the Gates of the Mountains country north of Helena, where Mann Gulch drains steeply west into the Missouri River. The crew flew in a Douglas C-47 and jumped in the late afternoon onto the slope opposite the fire. They collected their cargo after landing, and that is when they found the radio smashed.

Dodge went ahead to meet Harrison and look at the fire, and the crew started down the gulch toward the river on the side away from the flames. Dodge rejoined them and led them on down. At some point he saw that the fire had crossed to their side of the gulch below them, between the crew and the river. He turned the crew around and started them uphill toward the ridge through dry grass on a steep slope. According to survivor testimony as later reported, he told them to drop their heavy tools so that they could move faster.

With the fire closing from below, Dodge lit a fire in the grass ahead of him, let it burn out a patch, and lay down in the ash. The others did not join him. Sallee and Rumsey kept climbing and reached the ridge through a gap in the rock reef near the top. The fire overran the rest of the crew on the slope. The time usually given for the blowup is shortly before 6 p.m., drawn from the reconstruction in the Forest Service record and in Maclean’s book.

I am leaving out the minute-by-minute positions, the exact words Dodge used, and the individual exchanges that some retellings include. That material rests on testimony from three survivors who were spread across a burning slope, and the published accounts of what each man heard and saw do not agree in every detail. Anyone who needs that level of detail should read the testimony and the reconstructions directly and not rely on a summary, mine included.

One radio in one cargo chute: the link that broke on landing

Smokejumpers came down under personal parachutes, and the crew’s tools, food, water and radio were dropped separately in cargo bundles on their own chutes. The parachute on the radio’s bundle failed to open, and the set was destroyed on impact. The Forest Service accounts and Maclean both document the failure, and it is one of the few facts about the afternoon that nobody contests.

The radio was the crew’s connection to anyone outside the gulch. With it, the foreman could have told the dispatcher the crew was on the fire, reported what the fire was doing, asked for more people or equipment, and heard back whatever the dispatcher or anyone else had learned. The airplane returned to Missoula after the drop, so nobody was overhead to watch the crew, and as far as the published accounts show there was no plan for one. Once the cargo hit the ground, the crew could talk only to each other and to Harrison.

The record does not support the claim that a working radio would have saved the crew, even though that version is repeated often. Richard Rothermel’s 1993 Forest Service reconstruction of the fire’s behavior is titled “Mann Gulch Fire: A Race That Couldn’t Be Won,” and it describes a fire run up that grass slope faster than men climbing it once the blowup started. I have also not found an account showing that anyone outside the gulch saw the fire cross below the crew and was in a position to pass that warning in time. Because of that, the idea that a radio would have changed the outcome is speculation, however reasonable it sounds.

The record does support a narrower and more useful observation, which is about design. The crew had exactly one long-range link, and it traveled in the riskiest step of the operation, a free cargo drop into rough terrain, with no backup and no way of knowing whether it had survived until after the airplane was gone. That arrangement made the radio a single point of failure, and the failure happened at the first opportunity.

Tube sets and line of sight: what a 1949 crew could carry

The Forest Service had been building and fielding its own two-way radios since the 1930s, through an agency radio laboratory that designed sets for lookouts, ranger stations and field crews. My understanding of that equipment history is general rather than specific to this fire. The sets of the period ran on vacuum tubes and heavy batteries, and a crew set was enough of a load that a crew carried one and dropped it as cargo instead of strapping it to each jumper. I have not found a reliable published identification of the exact model the Mann Gulch crew carried, so I will not name one.

The handheld radio in the sense a firefighter means today did not exist in 1949. Bell Laboratories had demonstrated the transistor only in 1947, and years passed before transistorized two-way radios small enough for a belt or a chest harness were practical in field service. The Second World War had produced portable military sets, including the SCR-536 handie-talkie, but those were tube radios with limited range and battery life, and they were nowhere near standard issue for every member of a civilian fire crew. The model in which each firefighter carries a radio and hears the same traffic as the crew leader belongs to a much later generation of equipment.

Even a working crew set would have faced terrain. Mann Gulch is a steep side drainage off a river canyon, and a low-powered portable at the bottom of a gulch like that may or may not reach a lookout or a station, depending on where the far end sits. I have not seen a record of what this particular set could have reached from the crew’s positions, so I cannot say whether it would have worked from the slope even if it had arrived intact.

Once the radio was gone, the crew’s communications system was the human voice and line of sight between men walking in a file along a slope. That system works on a quiet hillside with a crew bunched together. It degrades quickly when the crew is strung out over a long distance, when wind and fire noise rise, and when smoke cuts visibility, and all of those conditions developed in the final minutes before the blowup.

An instruction with no frame: the escape fire nobody followed

Dodge’s escape fire is the moment in this story where voice communication most clearly broke down, and it is also the moment that later writers have examined hardest. By the survivors’ accounts as reported, Dodge lit the fire and called to the others to come into the burned area with him, and none of them did. Sallee and Rumsey later said, in substance, that they did not understand what he was doing. A foreman setting a fire in front of his crew while a larger fire chased them uphill looked to them like something other than a way out.

Karl Weick, an organizational psychologist, analyzed this moment in a 1993 article in Administrative Science Quarterly titled “The Collapse of Sensemaking in Organizations: The Mann Gulch Disaster.” Weick worked from Maclean’s book and the published record, and his conclusions are interpretation by a scholar rather than investigative findings. In paraphrase, his argument is that the crew was not a unit that had trained and worked together under Dodge, because smokejumper crews of that era were assembled from the roster for each jump, so the men had no shared history to tell them how to read their foreman’s unusual action. When the situation stopped making sense and the crew structure loosened as they spread out on the slope, Dodge’s instruction arrived with nothing to help them interpret it.

For communications people, this is the part of Mann Gulch with the most direct application. A transmission can be heard perfectly and still fail, because the receiver needs a frame to interpret it, and that frame has to be built before the emergency, in training and in working together. An instruction to do something counterintuitive, given under extreme noise and stress by a leader the crew does not know well, with no time to explain, is the hardest kind of message to get acted on. That holds on a fireground today just as it did on the slope in 1949, whatever radio is in the crew’s hands.

The Board of Review, Maclean, Rothermel and Weick: what each source is

The Forest Service convened a Board of Review soon after the fire and took testimony from the survivors and others. The board is generally described as having found no fault on the part of Dodge or the agency, and Maclean and later writers questioned how thoroughly it had examined what happened. The board’s record is the primary source closest to the event, and it reflects the questions an agency in 1949 thought to ask, which were not the questions fire behavior scientists or organizational researchers would ask decades later.

Norman Maclean, the author of A River Runs Through It, spent years on Mann Gulch, interviewing survivors, walking the ground and working with fire scientists. He died in 1990 with the book unfinished, and the University of Chicago Press published Young Men and Fire in 1992. The book mixes reporting, reconstruction and Maclean’s own reflection, and he says plainly where he is reconstructing. It is the most complete narrative of the fire, but its conclusions are one careful writer’s reading of the evidence and not an official finding.

Richard Rothermel was a Forest Service fire behavior researcher based in Missoula who worked with Maclean on the fire’s timing and spread. His 1993 Forest Service paper applies fire behavior modeling to the slope, the fuels and the weather to reconstruct how fast the fire moved compared with the crew. It is a technical analysis by a scientist, with the assumptions such modeling requires, and Rothermel states those assumptions. Weick’s article, discussed above, is a theory of organizational behavior that uses Mann Gulch as its case.

These four sources answer different questions, and quoting one of them as though it were another causes trouble. A sentence that treats Weick’s interpretation as a Board of Review finding, or Maclean’s reflection as Rothermel’s modeling, gives the reader a false idea of how the claim was established. The Forest Service’s historical collections and the Wildland Fire Lessons Learned Center are good places to begin for the primary material on historic fatality fires.

The 1957 task force: where the communications order came from

Mann Gulch was not the last fatal fire of that era. In 1953 the Rattlesnake fire in California killed fifteen firefighters, and in 1956 the Inaja fire, also in California, killed eleven. In 1957 the Forest Service chief appointed a task force to recommend ways to reduce the chances of firefighters being burned to death, and that group reviewed earlier fatality fires, with Mann Gulch commonly listed among them. Its best-known products are the Standard Firefighting Orders and the first list of situations that call for extra caution, which later became the Watch Out Situations.

One of the orders deals directly with the gap at Mann Gulch. As carried in the National Wildfire Coordinating Group’s Incident Response Pocket Guide, it reads: “Maintain prompt communications with your forces, your supervisor, and adjoining forces.” Another order requires leaders to give clear instructions and confirm they are understood, and others cover keeping informed on fire weather, knowing what the fire is doing at all times, posting lookouts when there is possible danger, and identifying escape routes and safety zones and making them known to the crew. Each of those orders corresponds to something the Mann Gulch crew lacked or could not do that afternoon. I want to be careful here, though: the task force drew on several fatal fires, so it would overstate the record to say that any single order was written because of Mann Gulch alone.

The Watch Out list carries the same concern in another form. It includes, in paraphrase, having no communication link with crew members or a supervisor and receiving instructions or assignments that are not clear. The orders and the Watch Out list have been revised and reorganized over the decades since 1957, and the wording in older training material does not always match the current version. Anyone teaching from them should use the current Incident Response Pocket Guide from NWCG and check the edition date before printing a handout.

LCES: how later doctrine tied communications to the escape route

LCES stands for Lookouts, Communications, Escape routes and Safety zones. It is commonly credited to Paul Gleason, a hotshot crew superintendent, who described it in the early 1990s as a compact way to keep the core safety elements in front of a crew. Its contribution was to present the four elements as one system in which each depends on the others, where the Standard Orders had listed them as separate items.

The dependency runs through communications. A lookout who sees the fire change direction or cross a drainage is useful only if that observation reaches the crew in time to act, and an escape route is useful only if the crew starts moving along it before the fire closes it off. The safety zone at the end of the route has to be identified and known to everyone, not just the leader, so that a crew scattered by smoke can still reach it. Without the communications link, a lookout becomes a witness instead of a warning, and the escape route becomes a plan known only to whoever made it.

Mann Gulch shows what the absence of all four looks like. No lookout was posted to watch the crew’s flank and the drainage below them, so nobody was in a position to see the fire cross. With the radio destroyed there was nobody to pass a warning to anyway. The escape route was improvised uphill once the danger was visible, and the only safety zone was the one Dodge burned for himself in the minute he had, which the crew did not recognize for what it was. LCES did not exist in 1949, but it describes those failures with precision. When fourteen firefighters died at South Canyon in 1994, LCES was already being taught, and how it was applied there became part of the review of that fire.

The conditions at Mann Gulch are not confined to smokejumping or to 1949. A brush company assembled from two volunteer departments, a strike team pulled together from several agencies for a single deployment, or a company working with one radio for the officer while the crew spreads out in tall grass all reproduce some of them: people who have not trained together, a thin or single communications link, and voice orders given across distance and noise. Today’s radios are better, and the human side of the problem has stayed much the same.

What to do at your agency

• The training officer pulls the current NWCG Incident Response Pocket Guide, compares the Standard Firefighting Orders and Watch Out Situations in it against the version in the department’s wildland training material, and replaces any outdated handout before the next wildland drill.
• The radio or apparatus officer walks every brush truck and wildland unit this month, records how many working portable radios each one carries and whether any crew member would be without a radio when the crew splits, and gives the list to the chief.
• The training officer runs one drill at an existing company training night in which the officer gives an instruction the crew has never practiced, and every member repeats it back before acting, so that the crew practices confirming an unfamiliar order before it ever faces a real one.
• The communications manager asks the state forestry agency for the communications plan it used on its most recent large fire in the county, and confirms that the channels in that plan are programmed in the department’s brush-unit radios.
• The battalion chief or senior officer adds one item to the next scheduled officers’ meeting: before any crew goes into grass or brush, the officer names the lookout, the escape route, the safety zone and the radio channel out loud, and the meeting minutes record that this is now the department’s practice.
• The training officer corrects any department training material that describes the Mann Gulch dead as “thirteen smokejumpers,” changing it to thirteen men, twelve smokejumpers and one fire guard, and adds the source type for each Mann Gulch claim on the slides.

Takeaways

• Thirteen men died at Mann Gulch on August 5, 1949, twelve of them smokejumpers and one a Forest Service fire guard, James Harrison, and Dodge, Sallee and Rumsey survived.
• The crew’s only radio was destroyed when its cargo parachute failed to open, which left the crew with no outside link for the rest of the afternoon.
• The record does not establish that a working radio would have saved the crew, and Rothermel’s 1993 fire behavior reconstruction describes a run the men could not outrun once it began.
• Equipment of 1949 meant one heavy tube radio per crew, dropped as cargo, and the practice of every firefighter carrying a radio came with much later technology.
• Survivor accounts indicate the crew did not understand Dodge’s escape fire, and Weick’s 1993 analysis attributes that in part to a crew that had not trained together under its foreman.
• The 1957 Forest Service task force, drawing on several fatal fires including Mann Gulch, produced the Standard Firefighting Orders, one of which requires prompt communications with your forces, your supervisor and adjoining forces.
• LCES, commonly credited to Paul Gleason in the early 1990s, treats communications as the link that makes lookouts, escape routes and safety zones work as a system.
• When citing Mann Gulch, name whether the claim comes from the Board of Review, Maclean, Rothermel or Weick, because each carries a different kind of authority.